Szczegóły publikacji
Opis bibliograficzny
Hypergraph grammar-based, multi-thread, multi-frontal direct solver scheduled in parallel GALOIS environment / Konrad JOPEK, Maciej PASZYŃSKI, Anna Paszyńska, Muhammad Amber Hassan, Keshav Pingali // Computer Science ; ISSN 1508-2806. — 2019 — vol. 20 no. 1, s. 27-55. — Bibliogr. s. 51-54, Abstr.
Autorzy (5)
- AGHJopek Konrad
- AGHPaszyński Maciej
- Paszyńska Anna
- Hassan M. A.
- Pingali Keshav
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 121988 |
|---|---|
| Data dodania do BaDAP | 2019-07-31 |
| Tekst źródłowy | URL |
| DOI | 10.7494/csci.2019.20.1.3010 |
| Rok publikacji | 2019 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp |  |
| Czasopismo/seria | Computer Science |
Abstract
In this paper, we analyze two-dimensional grids with point and edge singulari- ties in order to develop an efficient parallel hypergraph grammar-based multi- frontal direct solver algorithm. We express these grids by a hypergraph. For these meshes, we define a sequence of hypergraph grammar productions ex- pressing the construction of frontal matrices, eliminating fully assembled nodes, merging the resulting Schur complements, and repeating the process of elim- ination and merging until a single frontal matrix remains. The dependency relationship between hypergraph grammar productions is analyzed, and a de- pendency graph is plotted (which is equivalent to the elimination tree of a multi- frontal solver algorithm). We utilize a classical multi-frontal solver algorithm; the hypergraph grammar productions allow us to construct an efficient elimi- nation tree based on the graph representation of the computational mesh (not the global matrix itself). The hypergraph grammar productions are assigned to nodes on a dependency graph, and they are implemented as tasks in the GALOIS parallel environment and scheduled according to the developed de- pendency graph over the shared memory parallel machine. We show that our hypergraph grammar-based solver outperforms the parallel MUMPS solver. © 2019 AGH University of Science and Technology Press.
